Review for Biogeosciences

de la Vega et al: A δ11B-pH calibration for the high-latitude foraminifera species Neogloboquadrina pachyderma and Neogloboquadrina incompta

This study establishes a δ¹¹B–pH calibration for two key high-latitude planktonic foraminifera species N pachyderma and N incompta. These species are critical for reconstructing past ocean pH and atmospheric CO₂ levels in Arctic and subpolar regions, where existing records are sparse.

By using live-towed specimens collected from the Nroth Atlantic, and pairing them with in situ hydrographic measurements, the authors avoid common assumptions required in core-top approaches and find:

(1) Neogloboquadrina δ¹¹B values are lower than seawater δ¹¹B borate values. An interesting finding that is explained by localized acidification from respiration

(2) The δ¹¹Bforam to δ¹¹Bborate slope is greater than 1, a notable deviation from other non-spinose species calibrations. Authors suggest that vital effects may be enhancing pH sensitivity in these species.

(3) This calibration is applied to core-top samples from the Nordic Seas, demonstrating its paleoclimatic utility.

The motivation for the paper is sound and it is well-written. More studies like this are needed that explore the reasons why taxa can fall below the δ11B borate line and presumably are calcifying under pH conditions lower than ambient seawater. This work has implications for understanding boron isotope incorporation in other taxa with notably low δ11B (e.g. forams like H.elegans and bivalves). I would recommend the paper for publication following only minor edits.

Line by line points to consider:

Line 43: Cite Rae et al., 2018 Nature too here. A boron isotope paper that demonstrates the importance of carbon cycling and climate in the Southern Ocean

Line 64: Degrees Celsius doesn’t really need to be spelled out. Simply “°C” is unambiguous and widely understood.  

Line 108: This paragraph implies that only tow data are used when also core top data appear in the methods. Signpost the reader here to say that the calibration in this study is only composed of the tow data (for the reasons mentioned), and it is compared to core top data to demonstrate paleoclimatic utility.

Line 253: 0.5M HNO3 should be 0.5 M HNO₃ with a subscript. Check there is a space after M in other places where it is used in the text.

Line 258: both Ca isotopes were measured, but presumably only 1 was used for calculating ratios?

Presumably Mg/Ca and Sr/Ca reproducibility are considerable better than 5%

What [Ca] were samples run at? It says they were matrix matched, but to what concentration? 1mM?

Line 262: Stewart et al, 2020 Geostand Geoanal Res demonstrates the reproducibility of the Southampton method relative to other labs. This is important as the trace metal data in this study are from Southampton and AWI. It is important to demonstrate that consistent standard values were being hit by both labs so that the data can be compared directly without any interlab offset.

Line 266. New paragraph

Line 267. 1.1 permil is a large TPB correction. It would be useful to know which sample this applied to in the data table with a column stating the magnitude of TPB correction. The Y axis error bars are large in Figure 2. It might be worth reiterating what went into these propagated errors (i.e. it is more that analytical uncert). This is important are there are two particularly low δ11Bforam values (~11.5 ‰ ) that it could be argued are driving the slope >1 phenomenon. It would be good to show clearly that these are not samples that are badly impacted by TPB.

Line 270: repetition of class 100.

Line 280: Again, worth citing Stewart et al., 2020 here as these are bang on interlab consensus values for AE121 that was found to be 19.63 ± 0.17 ‰. i.e. when measured it is quite a bit lower than the original Vogel value.

Also, I think you mean MINUS 20.2 ‰ for the AE120, right? For the JCp value Gutjahr et al., 2020 should also be cited here.

Line 312: As above. The use of JCp-1 here can be used to discount interlab offsets in Mg/Ca between labs. Was the JCp cleaned or unclean? How does this value 3.94 compare to Hathorne et al., 2013 and also the measurements made in Southampton?

Line 320: add these core location to the map figure with labels.

Line 325: GLODAP should be all capital letters.

Line 327: modern pH/δ11Bborate is ambiguous (also line 368 – please check for other instances of this e.g. line 532). I think you mean “or” here, but it could be construed as the ratio of pH to δ11B borate.

Line 341: 432.6 μmol kg−1 needs a superscript. Otherwise it just means minus 1

Line 346: Does this monte carlo approach need a citation of Chalk et al., 2017?

Line 355: ok to group them for now, but it will be fascinating to see if N. pachyderma and N. incompta need to be separated for calibration purposes once more δ11B data are generated across a wide pH range for each taxon. The differences in Ba/Ca noted later in the text are intriguing and suggests they are living very different lives. I look forward to the next study on this!!

Line 388. I think this section 3.3 is what is needed in the intro to signpost the reader about what is to come before they see the methods.

Line 410: Could even bring coral data in here for comparison to show that this slope >1 is really unusual. You already use the Balanophyllia Gagnon 2021 citation elsewhere. Could also mention D.dianthus from Anagnostou 2012 and McCulloch et al., 2012.

Line 423: again “δ11Bforam-δ11Bborate” is an ambiguous dash that could be a minus sign

Line 456: Unsure what "privileged" is supposed to imply here. I’m not sure it is needed.

Line 471: CO3-2 should be “2-” not to the power of minus 2

Line 481: use words as part of a sentence rather than >> here.

Line 520: pH written twice

Line 565: space between units “50 m” and “150 m”. There are many other instances of this in the text. Please check.

Line 566: d11Bborate is now italic. Be consistent for clarity.

Line 601: “is” shouldn’t be italic

Line 672 “extent”

Table 1: add the tpb correction magnitude to this table. Also check the number of significant figures on the Ba/Ca column.

Table 2: define mswd

Figure 1: Caption says pCO2, but this is delta pCO2 relative to the atmosphere. This should be clearly explained in the caption. The figures could be better integrated with one another for instance plankton tows are red dots in Fig1 and therefore should be red dots in the data figure 2. Try to keep colours and symbols consistent to guide the reader’s eye of what is incompta and pachyderma and what is a tow or a core top. Use larger symbols for "this study" if you need to draw attention to what is new.

Figure 2: axis labels are ambiguous. The use of a dash (−) in “Foraminifera (tow) – δ11Bforam (‰) ” implies this is some value of foraminifera minus the δ¹¹B of calcite (same goes for the x axis). Y axis should read “δ¹¹B_foram (tow) [‰]” and x axis should read “Seawater δ¹¹B_borate [‰]”

Figure 3: more detail needed in the caption. Explain that this is a model.

Figure 4: again check for ambiguous minus signs. Do you actually mean minus in this case of d11Bforam – δ11B borate? It is not clear which taxa are plotted in panel b. the caption says pachyderma, but the extremely high Ba/Ca values suggest incompta are also on there. As per my earlier comment – figures need to be better integrated with the same colours and or symbols for tows and species where possible to guide the reader’s eye.

Figure 5: as above. Axis labels need revision for clarity. Caption needs to explain again PI and C_ANT so that the reader doesn’t need to search the main text.

Figure 6: y axis should read “…concentration [ppm]”. d18Oc should be δ¹⁸O_c. N pachyderma should be italic. Explain "CIAAN" Station. Again this could be better integrated with the map figure. e.g. Station 16 is mentioned often in the text but not highlighted on the map.

De la Vega et al presented a new calibration for boron isotopes in polar ocean planktic foraminifera Neogloboquadrina pachyderma/incompta using specimens from plankton tows. The authors explored why the slope of the calibration between d11B of foram and borate is larger than 1 and used new and published core top data to validate the new calibration.

The manuscript is well-written, and the new calibration is valuable to future paleo pH/CO2 reconstructions in the high-latitude oceans. 

The only major suggestion I have is to consider recalculating downcore pH/pCO2 from d11B reported in Yu et al 2013 and use this to further validate the new calibration.

Minor comments

L113: While I appreciate the advantage of a calibration based on plankton tows, I think it would be great to mention the caveat that the depths of the plankton tows are not necessarily the depth of calcification depths of the forams.

Fig 1: DpCO2 is shown in this figure. Calculated d11B of borate or pH would be better for the map.

Table 1: It would be great to have TA and DIC data in this table or in a supplementary table

Show raw and TPB-corrected d11B for all samples. Maybe also add a supplementary figure showing the calibration with raw d11B data to demonstrate that the large TPB correction (for some samples) is not driving the slope of the calibration.

L202-241: This part is mostly about deriving hydrographic data to calculate borate d11B. Consider moving these before L191 where the calculation of borate d11B was introduced

L205: The logic of constraining the calcification depth is not clearly described. I recommend adding a sentence stating that the calcification depth is determined by “comparing the d18Oc with equilibrium d18O in the water column”.

L210: N. pachyderma d18O, along with other data to decide calcification dept,h needs to be presented in a supplementary table

L261: new paragraph before “Samples for boron...”

L261: “boron [isotope] analysis”

L272: how the sigma is calculated is confusing. If the Anagnostou equation is employed, the sigma can be derived from 11B voltage and does not need to be derived from JCP-1 measurements.

L278: Some reorganization is needed here. It is essentially an interlab measurement comparison, but AWI measurements have not been introduced at this point.

L280: “-“missing for d11B of AE120

L302: Need to be reorganized. Triplicated measurements were mentioned before being introduced in L305.

L311: Unclear what 0.95+-0.47% is. It reads like this is the rsd for multiple El/Ca ratios, which would not be a proper way to report analytical errors.

L355: i.e. not e.g.

L360: including the number of data points will be good.

L364-367: unclear as written. Suggest rephrasing.

Fig 2: explain what errorbars are in panel 1 and what blue shades are in panels c,d.

X axes for panels A,B: borate d11B or d11B_borate

Y axis for panel B: foram d11B or d11B_foram

L383: there’s no fig. 2e

L449: not sure how applicable the Hönisch 2019 method is here. If ALK and DIC are lowered by 400 and 200 umol/kg, diffusion will elevate these concentrations. Maybe

L481: replace “>>”

L620: Sentence not complete